I think you may want to use -collapse- and outsheet to get the stats you
want, but I'm not sure what you mean by category_`X' in your commands -- do
you have a different categorical variable breakout for each of the 100
variables? Assuming that you dont,
collapse v1 - v100 , by(category)
will give you a dataset of means by category. Then
collapse (sd) v1-v100 , by(category)
will give you std devs. You can combine the two commands, but then you have
to type out new names for all 100 std dev vars.. Instead you can
use -renvars- to rename them after the fact and then use -merge- to combine
the means and std. devs. You may then want to re-order the vars and then
use -outsheet- to dump them to a text file.
Michael Blasnik
[email protected]
----- Original Message -----
From: "Camelia Minoiu" <[email protected]>
To: <[email protected]>
Sent: Friday, September 24, 2004 1:14 PM
Subject: st: tabstatmat repeatedly
> I would like to obtain summary statistics (by another variable) for a list
> of 100 variables, and then save all of these in one big matrix, and
> preferably have this exported into a text file.
>
> I have been following the discussion surrounding -tabstatmat-as well as
that
> regarding the command - mat2txt- (Feb 2004) but running mat2txt on my
> computer gives me the error message: "version 8.2 not supported" (I use
> Stata 8) and with tabstatmat I don't know if I can do this repeatedly. For
> example, writing a tiny program such as the one below does not seem to
work:
>
> foreach X of varist v1-v100 {
> quietly tabstat `X', statistics(mean, sd) by(category_`X') save
> tabstatmat mat`X'
> svmat mat`X'
> keep mat`X'*
> dropmiss, obs
> outsheet using c:\data\matrix.txt
> }
>
> Do you happen to know what I can do so that all the results from each
> tabstat on each variable are appended together in one big matrix?
>
> Thank you in anticipation.
>
> Camelia Minoiu
> PhD Student
> Department of Economics, Columbia University
>
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
